Add Markdown export to WAPI

[phenomen]

Currently WebAssembly build does not support Markdown export. I made this function available in WAPI.

Summary by CodeRabbit

• New Features

  • Added a Markdown export option to extract full document text as Markdown.

• Bug Fixes

  • Empty extraction results are now treated as "no data" — empty outputs return no content and length is set to 0.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds a new exported Zig function zpdf_extract_all_markdown(handle: i32, out_len: *usize) ?[*]u8 that returns full document text in Markdown, and updates the C API wrapper to treat zero-length extraction results as null; manifest line-count metadata updated.

Changes

Cohort / File(s)	Summary
New Markdown export (WASM API) src/wapi.zig	Added export fn zpdf_extract_all_markdown(handle: i32, out_len: *usize) ?[*]u8 — validates handle, retrieves document, calls doc.extractAllMarkdown(wasm_allocator), sets out_len, and returns the buffer pointer; preserves existing error and ownership semantics; returns null for invalid handle or missing/empty buffer.
C API: empty-result handling src/capi.zig	Updated wrapper to treat zero-length extraction as "no data": if allocated slice length is 0, free slice, set out_len to 0, and return null; otherwise return data pointer and length.
Manifest updates .../manifest	Reported lines changed: +22/-0 (manifest metadata adjusted).

Sequence Diagram(s)

sequenceDiagram
    participant Client as Client
    participant WASM as "WASM API\n(zpdf_extract_all_markdown)"
    participant DocStore as "Document Store\n(handle lookup)"
    participant Doc as "Document\n.extractAllMarkdown"
    participant Alloc as "WASM Allocator"

    Client->>WASM: call zpdf_extract_all_markdown(handle, out_len_ptr)
    WASM->>DocStore: validate handle & lookup document
    alt handle invalid or doc missing
        DocStore-->>WASM: null
        WASM-->>Client: return null (out_len = 0)
    else document found
        WASM->>Doc: call extractAllMarkdown(Alloc)
        Doc->>Alloc: request buffer
        Alloc-->>Doc: buffer ptr + length
        Doc-->>WASM: buffer ptr + length
        alt length == 0
            WASM->>Alloc: free buffer
            WASM-->>Client: return null (out_len = 0)
        else length > 0
            WASM-->>Client: return buffer ptr (out_len = length)
        end
    end

Loading

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐰 I nibble bytes and craft a sign,
I pull the markdown from each line,
Tiny hops through code and heap,
I bundle text and guard your sleep,
A rabbit's haul — concise and fine.

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'Add Markdown export to WAPI' directly and clearly describes the main change: adding Markdown export functionality to the WebAssembly API, which aligns with the PR objectives and file modifications.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@src/wapi.zig`:
- Around line 125-135: The zpdf_extract_all_markdown function can return a
pointer for empty buffers which breaks parity with
zpdf_extract_all/zpdf_extract_page; after calling
doc.extractAllMarkdown(wasm_allocator) (in the zpdf_extract_all_markdown branch
that handles documents[idx]), check if result.len == 0 and if so set out_len.* =
0 and return null, otherwise set out_len.* = result.len and return result.ptr;
preserve the existing guard for invalid handle and the final return null for
missing documents.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@src/wapi.zig`:
- Around line 122-142: The comment in zpdf_extract_all_markdown incorrectly
references toOwnedSlice while the function actually uses doc.extractAllMarkdown;
update the comment to describe that extractAllMarkdown returns an allocated
slice and that we treat zero-length results as "no data" by setting out_len to 0
and returning null; then make the C API (capi.zig) implementation consistent
with zpdf_extract_all_markdown by adding the same empty-buffer check around
extractAllMarkdown's result (check result.len == 0, set out_len.* = 0 and return
null) so both WAPI and C API handle empty slices identically.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@src/capi.zig`:
- Around line 194-200: The zero-length buffer returned by
doc.extractAllMarkdown(c_allocator) is allocated but dropped when the function
returns null, causing a leak; before returning null set out_len.* = 0 and free
the allocated slice via the same allocator (call c_allocator.free(result.ptr) or
the appropriate allocator.free for result.ptr) so the buffer is released; ensure
behavior matches zpdf_extract_all_reading_order and zpdf_extract_page_markdown
(or alternatively return result.ptr even for zero-length like those functions)
and retain the existing out_len handling and null return semantics used by
zpdf_free_buffer.